Agent evaluation fundamentals

Follow a sanitized mutating timeout from trace evidence through human review to an accepted eval definition—the point where Perseval stops today.

A failed trace is evidence from one execution. An eval requires a representative case, expected behavior, and a repeatable way to judge new runs.

A practical workflow for moving from recurring trace evidence to a reviewed eval definition, with the execution boundary kept explicit.

A useful comparison preserves both run identities, aligns equivalent work, and highlights the first meaningful divergence.

Ground truth records the best supported label for a case, while calibration measures whether graders and confidence estimates deserve trust.

White-box replay reruns agent logic while controlling recorded model, tool, time, or state dependencies.